1.2.1 Import des données exemple

-> Test d’import des donnèes “household” au format SPSS (.sav) :

library(haven)
  data_ex <- read_sav("model/zzhr62sv/ZZHR62FL.SAV")

1.2.2 Liste des variables (avec label)

# AFFICHER LE LABEL D'UNE VARIABLE
  attr(data_ex$HV028, "label")

[1] "Household weight for male subsample (6 decimals)"

# AFFICHER LES LABELS DE TOUTES LES VARIABLES (Code + label)
  l <- lapply(data_ex, attr, "label")
  l <- as.data.frame(l, stringsAsFactors = F)
  list_var <- as.data.frame(t(l))
  
  library(DT)
  datatable(list_var, options = list(
                                          autoWidth = TRUE, 
                                          rownames = FALSE,
                                          pageLength =NULL))

1.2.3 Liste des modalités (avec label)

# POUR UNE SEULE VARIABLE
  attr(data_ex[["HV023"]], "labels")

Region 1 - urban Region 1 - rural Region 2 - urban Region 2 - rural 
                 1                2                3                4 
  Region 3 - urban Region 3 - rural Region 4 - urban Region 4 - rural 
                 5                6                7                8 

# POUR TOUTES LES VARIABLES
  library(sjlabelled)
  get_labels(data_ex)

$HHID
   [1] "        1  1" "        1  2" "        1  3" "        1  4" "        1  5"
   [6] "        1  6" "        1  7" "        1  8" "        1  9" "        1 11"
  [11] "        1 13" "        1 14" "        1 15" "        1 16" "        1 17"
  [16] "        1 18" "        1 19" "        1 20" "        1 22" "        1 23"
  [21] "        1 24" "        1 25" "        1 26" "        1 27" "        1 28"
  [26] "        1 29" "        1 30" "        1 31" "        2  1" "        2  2"
  [31] "        2  3" "        2  5" "        2  7" "        2  8" "        2  9"
  [36] "        2 10" "        2 11" "        2 12" "        2 13" "        2 14"
  [41] "        2 15" "        2 16" "        2 17" "        2 18" "        2 19"
  [46] "        2 20" "        2 21" "        2 22" "        2 23" "        2 24"
  [51] "        2 25" "        2 26" "        2 27" "        2 28" "        2 29"
  [56] "        2 30" "        2 31" "        3  1" "        3  2" "        3  4"
  [61] "        3  5" "        3  6" "        3  7" "        3  8" "        3  9"
  [66] "        3 10" "        3 11" "        3 12" "        3 13" "        3 14"
  [71] "        3 15" "        3 16" "        3 17" "        3 18" "        3 19"
   [ reached getOption("max.print") -- omitted 6215 entries ]

  $HV000
  [1] "ZZ6"

  $HV001
  NULL

  $HV002
  NULL

  $HV003
  [1] "Incomplete household"

  $HV004
  NULL

  $HV005
  NULL

  $HV006
  NULL

  $HV007
  NULL

  $HV008
  NULL

  $HV009
  NULL

  $HV010
  NULL

  $HV011
  NULL

  $HV012
  NULL

  $HV013
  NULL

  $HV014
  NULL

  $HV015
  [1] "Completed"                                          
  [2] "No Household member/no competent member at home"    
  [3] "Entire Household absent for extended period of time"
  [4] "Postponed"                                          
  [5] "Refused"                                            
  [6] "Dwelling vacant or address not a dwelling"          
  [7] "Dwelling destroyed"                                 
  [8] "Dwelling not found"                                 
  [9] "Other"                                              

  $HV016
  NULL

  $HV017
  NULL

  $HV018
  NULL

  $HV019
  NULL

  $HV020
  [1] "All woman sample"    "Ever married sample"

  $HV021
  NULL

  $HV022
  NULL

  $HV023
  [1] "Region 1 - urban" "Region 1 - rural" "Region 2 - urban" "Region 2 - rural"
  [5] "Region 3 - urban" "Region 3 - rural" "Region 4 - urban" "Region 4 - rural"

  $HV024
  [1] "Region 1" "Region 2" "Region 3" "Region 4"

  $HV025
  [1] "Urban" "Rural"

  $HV026
  [1] "Capital, large city" "Small city"          "Town"               
  [4] "Countryside"        

  $HV027
  [1] "Not selected"     "Men's survey"     "Husband's survey"

  $HV028
  NULL

  $HV030
  NULL

  $HV031
  NULL

  $HV032
  NULL

  $HV035
  NULL

  $HV040
  NULL

  $HV041
  NULL

  $HV042
  [1] "Not selected" "Selected"    

  $HV044
  [1] "Household not selected" "Household selected"    

  $HV201
   [1] "PIPED WATER"                                         
   [2] "Piped into dwelling"                                 
   [3] "Piped to yard/plot"                                  
   [4] "Public tap/standpipe"                                
   [5] "TUBE WELL WATER"                                     
   [6] "Tube well or borehole"                               
   [7] "DUG WELL (OPEN/PROTECTED)"                           
   [8] "Protected well"                                      
   [9] "Unprotected well"                                    
  [10] "SURFACE WATER"                                       
  [11] "Protected spring"                                    
  [12] "Unprotected spring"                                  
  [13] "River/dam/lake/ponds/stream/canal/irrigation channel"
  [14] "Rainwater"                                           
  [15] "Tanker truck"                                        
  [16] "Cart with small tank"                                
  [17] "Bottled water or sachets"                            
  [18] "Other"                                               

  $HV202
   [1] "PIPED WATER"                                         
   [2] "Piped into dwelling"                                 
   [3] "Piped to yard/plot"                                  
   [4] "Public tap/standpipe"                                
   [5] "TUBE WELL WATER"                                     
   [6] "Tube well or borehole"                               
   [7] "DUG WELL (OPEN/PROTECTED)"                           
   [8] "Protected well"                                      
   [9] "Unprotected well"                                    
  [10] "SURFACE WATER"                                       
  [11] "Protected spring"                                    
  [12] "Unprotected spring"                                  
  [13] "River/dam/lake/ponds/stream/canal/irrigation channel"
  [14] "Rainwater"                                           
  [15] "Tanker truck"                                        
  [16] "Cart with small tank"                                
  [17] "Bottled water"                                       
  [18] "Other"                                               

  $HV204
  [1] "On premises" "Don't know" 

  $HV205
   [1] "FLUSH TOILET"                         
   [2] "Flush to piped sewer system"          
   [3] "Flush to septic tank"                 
   [4] "Flush to pit latrine"                 
   [5] "Flush to somewhere else"              
   [6] "Flush, don't know where"              
   [7] "PIT TOILET LATRINE"                   
   [8] "Ventilated Improved Pit latrine (VIP)"
   [9] "Pit latrine with slab"                
  [10] "Pit latrine without slab/open pit"    
  [11] "NO FACILITY"                          
  [12] "No facility/bush/field"               
  [13] "Composting toilet"                    
  [14] "Bucket toilet"                        
  [15] "Hanging toilet/latrine"               
  [16] "Other"                                

  $HV206
  [1] "No"  "Yes"

  $HV207
  [1] "No"  "Yes"

  $HV208
  [1] "No"  "Yes"

  $HV209
  [1] "No"  "Yes"

  $HV210
  [1] "No"  "Yes"

  $HV211
  [1] "No"  "Yes"

  $HV212
  [1] "No"  "Yes"

  $HV213
   [1] "NATURAL"                "Earth, sand"            "Dung"                  
   [4] "RUDIMENTARY"            "Wood planks"            "Palm, bamboo"          
   [7] "FINISHED"               "Parquet, polished wood" "Vinyl, asphalt strips" 
  [10] "Ceramic tiles"          "Cement"                 "Carpet"                
  [13] "Other"                 

  $HV214
   [1] "NATURAL"                  "No walls"                
   [3] "Cane / palm / trunks"     "Dirt"                    
   [5] "RUDIMENTARY"              "Bamboo with mud"         
   [7] "Stone with mud"           "Uncovered adobe"         
   [9] "Metalic sheets"           "Plywood"                 
  [11] "Cardboard"                "Reused wood"             
  [13] "FINISHED"                 "Cement"                  
  [15] "Stone with lime / cement" "Bricks"                  
  [17] "Cement blocks"            "Covered adobe"           
  [19] "Wood planks / shingles"   "Other"                   

  $HV215
   [1] "NATURAL"                 "No roof"                
   [3] "Thatch / palm leaf"      "Sod"                    
   [5] "RUDIMENTARY"             "Rustic mat"             
   [7] "Palm / bamboo"           "Wood planks"            
   [9] "Cardboard"               "Canvas/plastic sheeting"
  [11] "FINISHED"                "Metal sheets"           
  [13] "Wood"                    "Calamine / cement fiber"
  [15] "Ceramic tiles"           "Cement"                 
  [17] "Roofing shingles"        "Asbestos"               
  [19] "Other"                  

  $HV216
  NULL

  $HV217
  [1] "No adults"                "One adult"               
  [3] "Two adults, opposite sex" "Two adults, same sex"    
  [5] "Three+ related adults"    "Unrelated adults"        

  $HV218
  NULL

  $HV219
  [1] "Male"   "Female"

  $HV220
  [1] "95+"        "Don't know"

  $HV221
  [1] "No"  "Yes"

  $HV225
  [1] "No"  "Yes"

  $HV226
   [1] "Electricity"             "Lpg/Natural gas/Biogas" 
   [3] "Natural gas"             "Biogas"                 
   [5] "Kerosene"                "Coal, lignite"          
   [7] "Charcoal"                "Wood"                   
   [9] "Straw/shrubs/grass"      "Agricultural crop"      
  [11] "Animal dung"             "No food cooked in house"
  [13] "Other"                  

  $HV227
  [1] "No"  "Yes"

  $HV228
  [1] "No"                  "All children"        "Some children"      
  [4] "No net in household"

  $HV230A
  [1] "Observed"                           "Not observed: not in dwelling"     
  [3] "Not observed: no permission to see" "Not observed: other reason"        

  $HV230B
  [1] "Water not available" "Water is available" 

  $HV232
  [1] "No"  "Yes"

  $HV232B
  [1] "No"  "Yes"

  $HV232C
  [1] "No"  "Yes"

  $HV232D
  [1] "No"  "Yes"

  $HV232E
  [1] "No"  "Yes"

  $HV232Y
  [1] "No"                               "Yes: no cleansing agent observed"

  $HV234
  [1] "0 PPM (no iodine)"    "Below 15 PPM"         "15 PPM and above"    
  [4] "30 PPM"               "Salt not tested"      "No salt in household"

  $HV234A
  [1] "No iodine"            "Iodine present"       "No salt in household"
  [4] "Salt not tested"     

  $HV235
  [1] "In own dwelling"  "In own yard/plot" "Elsewhere"       

  $HV236
  [1] "Adult woman"                     "Adult man"                      
  [3] "Female child under 15 years old" "Male child under 15 years old"  
  [5] "Other"                          

  $HV237
  [1] "No"         "Yes"        "Don't know"

   [ reached getOption("max.print") -- omitted 7579 entries ]

1.5.1 Indicateurs pour le Bénin ?

Pour récuperer les indicateurs par enquête (date) disponibles pour le Bénin ainsi que leurs valeurs aggréger à un niveau infra-nationnal, il suffit d’utiliser la fonction dhs_data() :

# Make an api request
  resp <- dhs_data(surveyYearStart = 1990,
                   breakdown = "subnational",
                   countryIds = "BJ")
  
  str(resp)

'data.frame':   131923 obs. of  27 variables:
   $ DataId                : int  487479 3964044 489780 489783 489752 483472 487482 3964047 489778 489786 ...
   $ SurveyId              : chr  "BJ1996DHS" "BJ1996DHS" "BJ1996DHS" "BJ1996DHS" ...
   $ Indicator             : chr  "Age specific fertility rate: 15-19" "Age specific fertility rate: 15-19" "Age specific fertility rate: 15-19" "Age specific fertility rate: 15-19" ...
   $ IsPreferred           : int  1 1 1 1 1 1 1 1 1 1 ...
   $ Value                 : num  163 81 128 139 68 161 292 202 299 280 ...
   $ SDRID                 : chr  "FEFRTRWA15" "FEFRTRWA15" "FEFRTRWA15" "FEFRTRWA15" ...
   $ Precision             : int  0 0 0 0 0 0 0 0 0 0 ...
   $ RegionId              : chr  "BJDHS1996387001" "BJDHS1996387004" "BJDHS1996387007" "BJDHS1996387010" ...
   $ SurveyYearLabel       : chr  "1996" "1996" "1996" "1996" ...
   $ SurveyType            : chr  "DHS" "DHS" "DHS" "DHS" ...
   $ SurveyYear            : int  1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 ...
   $ IndicatorOrder        : int  11763010 11763010 11763010 11763010 11763010 11763010 11763020 11763020 11763020 11763020 ...
   $ DHS_CountryCode       : chr  "BJ" "BJ" "BJ" "BJ" ...
   $ CILow                 : chr  "" "" "" "" ...
   $ CountryName           : chr  "Benin" "Benin" "Benin" "Benin" ...
   $ IndicatorType         : chr  "I" "I" "I" "I" ...
   $ CharacteristicId      : int  387001 387004 387007 387010 387013 387016 387001 387004 387007 387010 ...
   $ CharacteristicCategory: chr  "Region" "Region" "Region" "Region" ...
   $ IndicatorId           : chr  "FE_FRTR_W_A15" "FE_FRTR_W_A15" "FE_FRTR_W_A15" "FE_FRTR_W_A15" ...
   $ CharacteristicOrder   : int  1387001 1387004 1387007 1387010 1387013 1387016 1387001 1387004 1387007 1387010 ...
   $ CharacteristicLabel   : chr  "Atacora/Donga" "Atlantique/Littoral" "Borgou/Alibori" "Mono/Couffo" ...
   $ ByVariableLabel       : chr  "" "" "" "" ...
   $ DenominatorUnweighted : chr  "" "" "" "" ...
   $ DenominatorWeighted   : chr  "" "" "" "" ...
   $ CIHigh                : chr  "" "" "" "" ...
   $ IsTotal               : int  0 0 0 0 0 0 0 0 0 0 ...
   $ ByVariableId          : int  0 0 0 0 0 0 0 0 0 0 ...

Nombre d’indicateurs disponibles par enquête pour le Bénin :

# Group by Survey and indicator
  temp <- aggregate(data = resp, Indicator ~ SurveyYearLabel + IndicatorId + IndicatorType, length)
  
  # Count Number of indicator by survey
  temp2 <- aggregate(data = temp, IndicatorId ~ SurveyYearLabel + IndicatorType, length)
  colnames(temp2) <- c("Year_of_survey", "Type_Indicator", "Number_of_indicators_available")
  
  # and plot the results
  library(ggplot2)
  ggplot(temp2, aes(x = Year_of_survey, y = Number_of_indicators_available, 
         fill = Type_Indicator)) +
    geom_bar(stat='identity')

Nombre d’indicateurs disponibles à chaque date pour le Bénin (depuis 1996) :

temp3 <- aggregate(data = temp, SurveyYearLabel ~ IndicatorId, length)
  temp3 <- temp3[temp3$SurveyYearLabel>=5,]
  
  # Nombre d'indicateurs présents à toutes les dates :
  print(nrow(temp3))

[1] 863

1.5.2 Donnèes aggrégées pour le Bénin ?

Les donnèes individuelles aggégées aux départements sont directement mis à disposition.

Exemple pour l’indicateur “Total fertility rate 15-49” (FE_FRTR_W_TFR), en 2012 :

resp2 <- dhs_data(indicatorIds = "FE_FRTR_W_TFR", 
                   surveyYearStart = 2012, 
                   surveyYearEnd = 2012,
                   breakdown = "subnational", 
                   countryIds = "BJ")
  
  
  datatable(resp2[, c("CharacteristicLabel", "Value")], 
            options = list(autoWidth = TRUE, 
                           rownames = FALSE,
                           pageLength = 18))

Le données sont également aggrégées par couple de département.

Pour récupérer les géométries du découpage administratif du Bénin directement depuis R, il est possible d’utiliser le package gadmr

if(!require(remotes)) install.packages("remotes")
  if(!require(gadmr)) remotes::install_github("SpatialWorks/gadmr")
  
  library(gadmr)
  # Téléchargement 
  benin_sp <- get_geopackage(country = "BEN",  layer = 1)

OGR data source with driver: GPKG 
  Source: "/tmp/RtmptvGtJZ/gadm36_BEN.gpkg", layer: "gadm36_BEN_1"
  with 12 features
  It has 10 fields

# Conversion objet sp to sf
  library(sf)
  benin_sf <- st_as_sf(benin_sp)
  
  # Chargement table de correspondance (à télécharger)
  table <- readRDS("table_passage_nom_ISO_Dep_Benin")
  
  # Merge avec la table de correspondance
  benin_sf <- merge(benin_sf, table, by.x="HASC_1", by.y="ISO" )
  
  # Merge avec les données
  benin_sf <- merge(benin_sf, resp2, by.x="NAME_dhs", by.y="CharacteristicLabel" )
  
  # Cartography
  library(mapsf)
  mf_theme("agolalight")
  mf_map(benin_sf, var = "Value", type = "choro", leg_pos = "left")
  mf_layout(title = "Total fertility rate 15-49, 2012",
            credits = "DHS program, 2012",
            scale = FALSE,
            arrow = TRUE,
            frame = TRUE)
  mf_inset_on(x = "worldmap")
  mf_worldmap(benin_sf)
  mf_inset_off()

2.3.1 Informations d’identification

config <- set_rdhs_config(email = "hugues.pecout@cnrs.fr",
                  project = "Summer School on Spatial analysis",
                  config_path = "~/.rdhs.json",
                  global = TRUE,
                  verbose_download = TRUE)

Répondre à la question par 1 ou 2 (puis “Entrée”) dans la console :

Une fenêtre pop-up s’ouvre pour saisir le mot de passe :

2.3.2 Créer le ‘coté client’

Créer le ‘coté client’ de l’API qui pourra être appelé comme argument dans différentes fonctions du package** avec client_dhs. Cela crée un cache pour toutes les datasets téléchargés.

client <- client_dhs(configuration)

2.3.3 Lister les datasets

La liste (et le nom de fichiers) des datasets accessibles est consultable depuis votre compte sur dhsprogram.com. Exemple :

Mais, il est également possible de lister tous les datasets (avec leur nom de code) de la manière suivante avec la fonction dhs_datasets

library(rdhs)
  
  # Séléction de l'ensemble des datasets sur le Bénin
  dat <- dhs_datasets(countryIds="BJ")

Cela génére un tableau dans lequel l’on peut repérer le nom des fichiers (FilName). Extrait :

2.3.4 Télécharger un dataset

Pour télécharger un jeu de données, utiliser get_datasets().

Exemple avec le dataset “Individual Recode” de 2027-2018 en format SPSS (.sav) :

# Ecrire le nom du fichier ciblé
  # Ou le numéro de ligne, ex : get_datasets(dat$FileName[167])
  DHS_BJ_data17 <- get_datasets("BJIR71SV.ZIP")

Pareil le dataset comportant les localisations GPS en 2017-2018 en format Flat ASCII (.dat) :

DHS_BJ_Geo17 <- get_datasets("BJGE71FL.ZIP")

Les donnèes sont alors téléchargées dans un fichier temporaire sur sa machine. Il ne reste plus qu’à les charger dans R.

2.3.5 Importer un dataset dans R

Utiliser la fonction readRDS()

# Données d'enquête prétéléchargées
  Data_BJ17 <- readRDS(DHS_BJ_data17$BJIR71SV)
  
  # Données géographiques prétéléchargées
  Geo_BJ17 <- readRDS(DHS_BJ_Geo17$BJGE71FL)

2.3.6 Visualiser les coordonnées GPS

Les données géographiques sont en format sp : Conversion en sf !

library(sf)
  Geo_BJ17 <- st_as_sf(x = Geo_BJ17)

Visualisation des données géographiques :

library(mapview)
  mapview(Geo_BJ17)

DHS recode -> https://dhsprogram.com/publications/publication-dhsg4-dhs-questionnaires-and-manuals.cfm

Documentation sur le traitement spatial des données :
- https://dhsprogram.com/pubs/pdf/SAR11/SAR11.pdf

Documentation sur prevR package :
- https://cran.r-project.org/web/packages/prevR/index.html
- https://journals.openedition.org/cybergeo/24606
- https://cran.r-project.org/web/packages/prevR/vignettes/intro_prevR.html